Selenium (Se) has shown cancer preventive activity in a long-term NCI sponsored human intervention trial in which a daily supplement of 200 mg was given to a population with a daily intake of less than 100 mg, similar to that of much of the U.S. population. Only limited information is available on the baseline (unsupplemented) metabolism of either inorganic or organic Se in humans, and no information is available on changes in metabolism of either form that would likely result from a three-fold increase in intake. A study is in progress to provide information on the pharmacokinetics of Se in its prototype forms: sodium selenite and selenomethionine (SeMet). This information is necessary for the determination of time and manner of administration in future human trials. Integrated kinetic models are being used to interpret the study data. Various body pools have been hypothesized and rates of exchange between them estimated, as well as residence times. The models indicate important kinetic differences between selenite and selenomethionine. Alternative models were used to investigate one of the most important differences, the recirculation of the organic, but not the inorganic form. The models have been modified and combined into a single model to better simulate dietary intake of selenium. Model parameters have been reestimated for the selenite model, taking into account body stores of tracer present at the time the dose was given in the pharmacokinetics study and of tracer present in the diet. The introduction of tracer extraneous to the dose resulted in substantial changes to many model parameters. Estimates of body stores and of half-lives exceed previous estimates. These results were presented at an international Se symposium. In collaboration with the Cancer Prevention Studies Branch; the Beltsville Human Nutrition Research Center, USDA; and Cornell University, a second kinetics study to last three years has been designed, which will include 1) a 6-month baseline study, in which data on both inorganic and organic Se will be collected to permit refinement of the above models; 2) a 2-year period in which subjects will be supplemented with 200 mg/d of SeMet; and 3) a second 6-month kinetics study. Possible changes in metabolism, in the distribution of plasma seleno-proteins and in the composition of the colonic microbial community arising from long-term selenium supplementation will be investigated.